Use of AlignRT InBore in a Lung SABR Workflow on a Ring Gantry Linac

Hannah Nayee (00:04):

So yeah, we’re at the Royal Surrey Cancer Center and we’re going to talk to you about how AlignRT InBore works with the Ring Country linac for our lung SABR patients. So just to introduce us, we are a two-centre department. We’ve got our main center in Guilford and then our satellite center in Red Hill. And in our satellite center in 2024, we removed our TrueBeam’s and we installed two Halcyon machines. One that was upgraded to ethos and then one that’s kept as Halcyon. In our main center we have four TrueBeam’s and one Ethos linac. The ones at the main center do not have any SGRT, though we will be getting some soon. And our satellite center was equipped with HyperSight and also the Inor surface guided system. So you can see in the image there, you’ve got your horizon cameras on the ceiling, which are your out of bore cameras, and then the patient goes into the ring of the bore, the bore of the machine and there’s a ring inside which tracks them when they’re in their treatment position.

Hannah Nayee (01:13):

So the SABR service was moved to the main site during our replacement of the machines. So it was really important for these patients to be able to walk back to the main center as soon as possible, as it was causing capacity issues for our main site. And the patients were travelling quite far for their treatment. So having them back at the satellite center has been really positive for these patients, but there isn’t much data from UK centers using SGRT and also doing SABR on their ring gantry machines. So we created an implementation working party made up of MTT the physicists, engineers, radiographers and also operational teams working together to get this project underway. We needed to look at what the planning looked like for these patients, as we have a ring entry at the main site, but we’d never tried SABR there, mainly because the cameras that come with a healthy scene and they don’t allow you to look inside the bore of the machine and actually see if the patients are moving.

Hannah Nayee (02:14):

So it’s quite essential if you have SGRT for these patients. We needed to look at the staff training. So the staff at the satellite center hadn’t treated any SABR patients for over a year. So when it came to implementing these patients, again, we needed to send these staff over to the main site for some refresher training, or we had staff that had left, and our skill mix was a bit low on the SABR side, so we needed to get more staff trained up in the SABR. The housing doesn’t have 4D CBCT capabilities, so in our department, we have a protocol where you need to have 4D CBCT taken either at day zero or fraction one for the patients. So we amended our workflow, and I’ll talk a bit more about that afterwards. And then of course we wanted to look like look at what SGRT monitoring would look like for these patients. Were the ROIs going to be the same as our thorax patients? Did we need to amend these? So we started with looking at the first 10 patients of where we used AlignRT and how we monitored them.

Hannah Nayee (03:18):

So a little bit about the planning study that we did, as I think it’s quite useful for anyone who’s looking at implementing this. We inquired with Varian, and we were told that you can’t switch off the automation of the beams when treating patients on the Halcyon. So if anyone doesn’t have one should know that you take your CBCT scan, and then it will automatically order your arcs in how you want to treat them. But if you are taking an interim scan, like how we do on our saver patients, you don’t have much of a leeway in time to stop the beam and that’s all radiographer dependent. So we had a look at a few different plans, tested them out on the machine. The first plan was with one isocenter and two arcs with the same gantry and collimators for both arcs.

Hannah Nayee (04:05):

And that only allowed for a two-second pause for the radiographers to pause that first arc beam and take the CBCT. The second plan was similar, but there was a gantry and collator twist between the arcs, which allowed for slightly more time. But I’m sure, as you know, our control rooms get many distractions, a rogue patient walking in, an operation manager coming down for a chat, and it’s that if that time’s gone, you’ve then gone into your second arc and you’ve missed the opportunity. So we really wanted to make sure that there was something robust that would make sure that we would, you know, be able to take that interim scan and reduce that radio for error. So our planning team came up with a solution of having two isocenters, which had a very small shift of 0.25 centimeters in between, it didn’t cause much of a change for the planning team and also it meant that the Halcyon would stop after the first arc and then it would prompt the user to motion enable and move to the second one.

Hannah Nayee (05:02):

At that point, that prompts a radiographer to go back to the CBCT field, take the interim scan and then you can treat your second arc. So that was considered the safest option. Then, moving on to our treatment verification workflow, so per SABR consortium guidelines need to take a 4D CBCT scan on these patients. So we don’t have any TrueBeam’s at the satellite center. So our only option was to send these patients to the main site for a day zero assessment. If their tumor motion is within our department tolerance, then they would continue to come back to the satellite center and have their treatment on a ring gantry linac. If we didn’t, if the patient’s tumor motion was more than the protocol tolerance, then they would have a TrueBeam plan made and then be treated at the main site. Our patients are prompted of this; they’re aware of what the sort of process is and of course, they just want the most accurate treatment at the time, so they’re not too fussed about what happens at the end of it.

Hannah Nayee (06:03):

And it is only about five to eight treatments that they would be going to the main site and that’s what’s made it work for us. So the main audit objectives for our SGRT audit were to evaluate how the SGRT works with the ancillary equipment, such as the abdominal belt compression. We were concerned whether that would get in the way of the cameras and any ROIs. Also, compare the use of the monitoring to the CTs and how accurate that was for actually the tumor motion and what we saw even in the interim scans and then to evaluate how reliable the SGRT monitoring was for the real-time monitoring for these patients, hand over to Gail.

Gail Anastasi (06:51):

So just,

Speaker 3 (06:53):

Yeah, the hierarchy, it’s clicker.

Gail Anastasi (06:55):

Okay. Okay. So yeah, so the method we used the patients as Hannah explained, were planned with the two isocenter with a 2.5 millimeter shift between the sup in the SUP/INF direction and generally we found that in terms of planning the they were very similar in terms of monitor units and our conformity initially we double planned, so we had sort of a TrueBeam planned as a backup and we had our Halcyon team plan and we didn’t really see any difference. I mean, you could argue that by doing our 2.5 millimeter shift you introduce sort of a half millimeter sort of beam width, but actually on SABR plans on certainly for lung, which are spherical in nature, we didn’t really see this improved sort of conformity maybe on more complex plans. Potentially, you could. And the other thing to keep in mind I guess, but this is for I guess all SGRT is that for very post tumors, we sometimes had to have an isocenter maybe that wasn’t at the center of the tumor to ensure that your sort of region of interest at the surface is not outside your field of view for your SGRT.

Gail Anastasi (07:58):

So just sort of to keep that in mind in terms of immobilization, we just followed our standard immobilization for SABR lungs. We had arms up on a wing board with a vacuum bag for arm support. Our patients with middle to lower lobe tumors are typically treated with abdominal belt compression. And as Han explained, we looked at appropriate treats of interest for these for these patients.

Gail Anastasi (08:28):

And then for each patient we recorded real time, the real time delta shifts for vert, long, lat,  your roll and pitch from the InBore SGRT system. And we did this at specific time points. So looking at these points, the first point was before the first cone beam CT acquisition. The second point was before applying the cone beam CT moves and that was to look to see if there was, there were any sort of shifts or whilst we were matching the cone beam because that initial cone beam sometimes takes a while to match. And then we looked at the first arc delivery to see whether during the arc delivery, there were any, any moves. We then again looked after our interim cone beam CT and finally after our second arc delivery. So those were the five time points that we recorded our deltas. And for consistency and reliability, we always recorded delta at exhale, so you can, you can sort of see, see the trace.

Gail Anastasi (09:30):

And so there was a radiographer, so we had a sort of proforma, like a form with all the time points per patient, and we did have a radiographer there sort of writing down the deltas at each of the time points. And then we also looked at the deltas from the online cone beam CT match. For comparison, I’ll explain this later. So in terms of our results, we had 10 patients in the audit, and these included eight upper lobes and two middle lobes. The region of interest that we chose excluded the abdominal belt and these were, so we found that a standard thorax region of interest was not necessarily for our SABR patients. So in the first image you can see what our sort of standard ROI tends to look like. And then the second is sort of what we tend to use for an upper lobe. And then the third sort of for something a lower lobe and obviously for some of our SABR treatments we’re treating say more than one lesion at a time. So we might have several regions of interest and our delta tolerances that we had were 0.2 centimeters for our vert, long and lat and three degrees for our roll and pitch.

Gail Anastasi (10:50):

So looking at this in a bit more detail, the mean difference between did I skip, sorry, apologies. Yeah, I did, sorry, <laugh>. So across all our 10 mean cha across all our 10 patients, we looked at the change in deltas during the time the cone beam CTs were being matched and during the arc delivery. So what we can see in the graph here on the Y axis we have the mean delta in centimeters or degrees and then across the X axis we have sort of little groups of say the vert during the cone beam CT matching and the vert changed during the arc. And then similarly for long, lat, your roll and pitch and we saw that the vert, long and lat mean change was less than a millimeter under your roll and pitch was less than 0.3 degrees. And you can see maybe the spread of results is slightly shorter for the arcs. And I think that’s because the arc treatments are quite quick with triple F and slightly longer longer time taken for cone beam CT matching. But generally even within that spread the results are less than sort of 0.25 centimeters for vert long and, and lat, and yeah, the in and the euro and pitch, we felt that these were sort of adequate for our SBRT lung treatments. It gave us sort of confidence in our setup there.

Gail Anastasi (12:25):

Then as I said, we looked at the difference between the AlignRT deltas and our cone beam CT soft tissue match and we found that there was no significant sort of difference there. And this was quite reassuring I think. Comparatively, when we set up two tattoos on our TrueBeam’s and then we do a soft tissue match, we often do see sort of moves of three sort of four moves that, you know, we have to make. Whereas I think when we’re setting up with SGRT, we less, but when we do the soft tissue match tends to agree a lot, a lot better. So I think we are getting better setups to start off with.

Gail Anastasi (13:00):

And then most importantly, perhaps when AlignRT did flag a move, so we did see like a delta come out with this justified an additional image for us and we did see a soft tissue positional change that was confirmed. So, so that, that’s quite reassuring. And in terms of time, we found that patients experience shorter treatment appointments by an average of nine minutes. So our average treatment time was about 21 minutes. We generally have a half-hour treatment slot for our long SABR. So we were doing quite well, and we can see the range there. I have to say that this can’t be attributed to treating on a Halcyon or SGRT because we did implement the two together. So it’s a combined result, there in a shortening of the treatment time.

Gail Anastasi (13:53):

So in conclusion, we feel that IPO, SGRT for our lung SABR was adopted easily with a decreased treatment time on the treatment couch. And to date, we’ve treated more than 30 patients. We felt that the sub-millimeter change in deltas during arc delivery makes us confident that we have a safe treatment approach. As I said, there was no significant difference between the AlignRT deltas and what we saw in our cone beam CT soft tissue match. And therefore, because of this we feel moving forwards that we might consider omitting the interim cone beam CT and only acquiring if the AlignRT deltas are out of tolerance. We haven’t done this yet, but it’s sort of something we, we think about moving forwards. So finally I’d like to acknowledge all the staff as Hannah said this, and as me mentioned before, these are multidisciplinary efforts between the radiographer physicists, our engineering team, and ops teams. And this has allowed us to establish this at our satellite center now a standard of practice which is great for our patients. And yeah, thank you for listening, and we are happy to take any questions. Thank you.